We Don't Know What to Eat
How bad science created a misinformed national diet—and did nothing to slow the growth of obesity.
If you go to the National Institute of Health’s website today, you will find a section on a "Healthy Eating Plan.” That plan recommends a diet “low in saturated fats, trans fat, cholesterol, salt, and added sugars, and controls portion sizes.” These recommendations may well have been copied and pasted from 1977.
Nothing has changed over the past 36 years, except for this: everyone is fatter.
The U.S. government began issuing dietary guidelines in 1977, when the Senate Select Committee on Nutrition and Human Needs, led by Senator George McGovern, issued the first dietary recommendations for the American people. Although these recommendations were made some 36 years ago, you probably recognize them immediately: “Increase consumption of complex carbohydrates and ‘naturally occurring sugars;’ and reduce consumption of refined and processed sugars, total fat, saturated fat, cholesterol, and sodium.” And those should sound identical to your doctor’s advice: decreased consumption of refined and processed sugars; foods high in total and animal fat, eggs, butterfat, and other high-cholesterol foods; and foods high in salt.
There’s little to no good science behind our diet.
According to the Centers for Disease Control, obesity has more than doubled among adults since these dietary recommendations were put in place in the 1970s, and as of 2010, more than one-third of Americans were obese. Over the same time, the rate of diabetes has quadrupled, up to eight percent of the population in 2011. Clearly, something hasn’t been going according to plan.
Perhaps you’ve witnessed someone struggle with a diet, or struggled yourself. It’s not just stuff of TV shows; people breaking down, sobbing, wishing they looked differently and trying incredibly hard but it just isn’t working. This happens to real people, millions of them. It seems odd and a bit heartless to assert that this meteoric rise in obesity and associated diseases is a result of people not trying hard enough.
But there’s another explanation, one that’s gaining traction across the scientific community. Maybe the science behind this diet was bad, and the decision to launch the country into the diet was a poor one, and the non-decision to back off in the face of contradictory evidence even worse. At its most charitable, these experts say, it was a bad experiment. At its worst, it was a crime that has cost millions of lives, and the toll keeps rising.
THE SENATE SELECT COMMITTEE on Nutrition and Human Needs based their recommendations largely on the Seven Countries Study, which was first published in 1970 and led by University of Minnesota researcher Ancel Keys, whose findings were affirmed by several subsequent, large-scale studies such as the Nurses’ Health Study, which found that high saturated-fat diets were related to high cholesterol, and higher cholesterol in turn led to higher risks of obesity, heart attack, stroke, heart disease, and mortality. The Seven Countries Study painted a direct link between dietary fat, misery, and death—and that’s been the story ever since.
But there were issues from the start.
“Keys chose seven countries he knew in advance would support his hypothesis,” Gary Taubes wrote in Good Calories, Bad Calories: Fat, Carbs, and the Controversial Science of Diet and Health. “Had Keys chosen at random, or, say, chosen France and Switzerland rather than Japan and Finland, he would likely have seen no effect from saturated fat, and there might be no such thing today as the French paradox—a nation that consumes copious saturated fat but has comparatively little heart disease.”
Zoe Harcombe, author of the Obesity Epidemic: What Caused It? How Can We Stop It?, also found, using World Health Organization data, that not only is there no statistical correlation between mean cholesterol levels and mortality, but there’s no positive relationship whatsoever.
“Cholesterol (and protein and phospholipids and triglyceride—the four substances found in all lipoproteins) is found at the scene of damage to arteries,” Harcombe told me, “but the four vital components of lipoproteins are there to repair that damage. They did not cause the damage any more than police caused the crime when they are found at the scene of that crime.”
According to Harcombe and Taubes, Keys used cherry-picked data to reach a logically-flawed conclusion, but it was the biggest study available, so George McGovern jumped on it because, in his words, “Senators don’t have the luxury that a research scientist does of waiting until every last shred of evidence is in.”
But what about the studies that affirmed the Seven Countries research?
FOUNDED IN 1976—A year before McGovern’s recommendations—the Nurses’ Health Study takes surveys of nurses’ health habits. These types of studies—including the Seven Countries Study—are called “observational studies,” and they can only tell us so much. The first Nurses’ Health Study followed 121,700 nurses between the ages 30 and 55 between 1976 and 1989, a massive sample that is sure to capture a wide variety of individuals. But the conclusions ignore this and instead focus on individual effects, even though the participants were free to live their lives as they wanted during the 13 years of the study. Nothing was controlled; all health-related variables were in play.
Taubes outlined one of the chief issues with such a study in a 2007 New York Times Magazine article, which he quoted in this blog post. It’s known as the “compliance effect.”
Quite simply, people who comply with their doctors’ orders when given a prescription are different and healthier than people who don’t. This difference may be ultimately unquantifiable. The compliance effect is another plausible explanation for many of the beneficial associations that epidemiologists commonly report, which means this alone is a reason to wonder if much of what we hear about what constitutes a healthful diet and lifestyle is misconceived.
This Nurses’ Health Study, then, is only really telling us who leads a healthy lifestyle and who doesn’t.
The worst of it is, we still make these elementary mistakes. In 2012, a study was released that supposedly affirmed red meat’s link to death, cancer, and heart risk. When Harcombe looked at the actual data, she found the same thing as the Nurses’ Health Study: correlations that simply don’t tell us anything. Some excerpts from her analysis:
• “As red & processed meat consumption increases, so exercise falls. Could lack of exercise impact mortality?”
• “As red & processed meat consumption increases, so does BMI. Could BMI impact mortality?”
• “As red & processed meat consumption increases, so does smoking – the top quintile virtually three times higher than the lowest. Could smoking impact mortality?”
• “As red & processed meat consumption increases, so does diabetes. Could diabetes impact mortality?”
• “As red & processed meat consumption increases, so does calorie intake. Could calorie intake impact mortality?”
• “As red & processed meat consumption increases, so does alcohol intake. Could alcohol intake impact mortality?”
So instead of possibly linking exercise, Body Mass Index, diabetes, smoking, caloric intake, or alcohol intake to mortality, the conclusion was that, no, it is red meat that impacts mortality. It’s the compliance effect, again. To isolate red meat as the culprit is to ignore variables the researchers were not controlling for. It is, in short, bad science.
“The Nurses Health Study showed exactly the same correlations—the numbers were slightly different but the trends were the same,” Harcombe wrote in her analysis of the 2012 red meat study. “As red and processed meat consumption increased so exercise and high cholesterol fell; BMI, smoking, diabetes, calorie intake and alcohol intake all increased.”
Furthermore, according to the Nutrition Science Initiative, a foundation co-created by Taubes to yield better science behind epidemiology, the purported results from these landmark studies have never been consistently replicated in controlled environments. There’s little to no good science behind our diet.
So if our recommended diet is faulty, what should we eat? This is where the real harm of the last 35 years of questionable science comes to the forefront: we simply don’t know yet. When you spend the better part of three decades chasing a ghost, all you’re left with is a pretty good idea that there is no ghost. The medical community’s dedication to these established diets had led us burrowing deeper into the same rabbit hole, rarely exploring new pathways.
We have to eat, though, so when asked for dietary advice, experts need to say something. And that brings us right back to theories.
As he wrote in Why We Get Fat: And What To Do About It, Taubes believes a high-fat, moderate protein diet is the best one, because insulin triggers hormones that put fat in our fat tissue, and a bit ironically, fat is the one nutrient that doesn’t trigger insulin secretion. Harcombe told me the root of a good diet is avoiding foods that didn’t exist before the obesity epidemic. Or: “eating real food. Meat, eggs and dairy foods from pasture living animals; fish; vegetables; salads; nuts/seeds; fruits in season—that's the basis of a good diet.” You’re likely to encounter other diets that purport to have the answers as well. They may or they may not, but at least we can be pretty sure of one diet that doesn’t work. It only took us three decades and an epidemic to prove it.